U.S. patent application number 10/317555 was filed with the patent office on 2004-06-17 for hand-held tester and method for local area network cabling.
Invention is credited to Davis, Keith H., Pivonka, Ed, Renken, Gerald W..
Application Number | 20040113604 10/317555 |
Document ID | / |
Family ID | 32325946 |
Filed Date | 2004-06-17 |
United States Patent
Application |
20040113604 |
Kind Code |
A1 |
Renken, Gerald W. ; et
al. |
June 17, 2004 |
Hand-held tester and method for local area network cabling
Abstract
A LAN tester has display and remote units each having a
connector jack attached to an adapter board for connection to the
plug of a patch cord. Both the display and remote units have
circuits which are capable of measuring the phase between a drive
signal voltage and the corresponding coupled or reflected signal
due to the drive signal. Scattering parameters for the mated
connector pairs and the patch cord itself are measured during a
field calibration. A computer in one or both of the tester units
stores the measured scattering parameters and uses the scattering
parameters to move the reference plane to any desired location
along the patch cord. Channel link or permanent link tests can be
conducted using the same equipment.
Inventors: |
Renken, Gerald W.;
(Carlsbad, CA) ; Davis, Keith H.; (San Diego,
CA) ; Pivonka, Ed; (Rancho Santa Fe, CA) |
Correspondence
Address: |
Joel H. Bock
COOK, ALEX, McFARRON, MANZO,
CUMMINGS & MEHLER, LTD.
200 West Adams Street, Suite 2850
Chicago
IL
60606
US
|
Family ID: |
32325946 |
Appl. No.: |
10/317555 |
Filed: |
December 12, 2002 |
Current U.S.
Class: |
324/76.52 ;
324/543 |
Current CPC
Class: |
H04L 41/26 20130101;
H04L 43/50 20130101 |
Class at
Publication: |
324/076.52 ;
324/543 |
International
Class: |
G01R 023/00; G01R
023/12 |
Claims
We claim:
1. A LAN cabling testing system, comprising: first and second patch
cords each terminating at first and second plugs; a hand-held
display unit and a hand-held remote unit, each one of said units
including means for sending and receiving a wave form of selected
frequency to and from the other of said units through said patch
cords and a LAN link to be tested; the hand-held display unit
including a jack for receiving a plug of one of the patch cords,
said jack and plug defining a first mated connector pair; the
hand-held remote unit including a jack for receiving a plug of the
other of the patch cords, said jack and plug defining a second
mated connector pair; and phase measuring means for measuring phase
in one of the display or remote units.
2. The LAN cabling testing system of claim 1 wherein one of the
display unit and remote unit further comprises electronic storage
means for storing the scattering parameters of the patch cords and
the mated connector pairs, and calculating means programmed to use
the stored scattering parameters to move the phase reference planes
to the necessary locations for the type of test desired.
3. In a LAN cabling testing system of the type having a display
unit and a remote unit and first and second patch cords of known
lengths L1 and L2, the patch cords each terminating at first and
second plugs, and the display and remote units each having a jack
for receiving a patch cord plug, a plug and jack when connected
comprising a mated connector pair, the display and remote units
each having means for sending and receiving a wave form of selected
frequency to and from the other unit through said patch cords and a
LAN link to be tested, an improved method of testing LAN cabling
comprising the steps of: a) calibrating the patch cords and mated
connector pairs, calibration including the step of measuring the
scattering parameters of the mated connector pairs and the patch
cords; b) connecting the first patch cord to the display unit and
one end of the link to be tested, and connecting the second patch
cord to the remote unit and to the other end of the link to be
tested and shooting the link to be tested; and c) using the
scattering parameters of the mated connector pairs and the patch
cords to move the reference planes at the display unit and remote
unit to the necessary locations for the type of test desired.
4. The method of claim 3 wherein the calibration step further
comprises the steps of: a) connecting the first plug of the first
patch cord to the display unit's jack and the second plug of the
first patch cord to the remote unit's jack; b) measuring all four
scattering parameters ST1 of the first patch cord including the
mated connector pairs at each unit; c) repeating steps a and b for
the second patch cord to obtain the scattering parameters ST2 of
the second patch cord including the mated connector pairs at each
unit; d) obtaining values for frequency F of the waves generated by
the units, the nominal velocity of propagation NVP, and the
attenuation constant .alpha. and calculating the scattering
parameters SB1 and SB2 of each patch cord without its plugs using
its known length and the obtained values of F, NVP and .alpha.; e)
calculating the scattering parameters SA1 and SC1 of each mated
connector pair of the first patch cord from ST1 and SB1; and f)
calculating the scattering parameters SA2 and SC2 of each mated
connector pair of the second patch cord from ST2 and SB2.
5. The method of claim 4 further comprising the step of saving the
total measured scattering parameters ST1 and ST2 and the calculated
scattering parameters SA1, SB1, SC1, SA2, SB2 and SC2 in one of
said units.
6. The method of claim 4 wherein the step of obtaining values the
nominal velocity of propagation NVP, and the attenuation constant
.alpha. is performed by measuring these quantities.
7. The method of claim 4 wherein the step of obtaining the nominal
velocity of propagation NVP, and the attenuation constant a is
performed by assuming appropriate values of these quantities.
8. A LAN cabling testing system, comprising first and second patch
cords each terminating at first and second plugs, a hand-held
display unit and a hand-held remote unit each including a channel
link adapter card having a jack suitable for receiving a plug of a
patch cord, each one of said units including means for sending and
receiving a wave form of selected frequency to and from the other
of said units through said patch cords and a LAN link to be tested,
the display unit and remote unit including phase measuring means
for measuring phase in one of the display or remote units such that
channel link and permanent link tests can be made using the channel
link adapter card.
Description
BACKGROUND OF THE INVENTION
[0001] Local area network (LAN) cabling is used to connect
equipment such as personal computers, printers and fax machines
that pass information between them using high-speed digital
signals. This type of high performance cabling is sometimes
referred to as telecommunications cable. Since an office contains
many computers, computer file servers, printers, and fax machines,
the LAN cabling interconnects all of this equipment into a
communications network. LAN cabling has been designed to support
telecommunication between all of the individual elements of the
network.
[0002] FIG. 1 shows an example of LAN cabling, in a simplified
drawing. FIG. 1 shows how the LAN cabling, most of which runs
within the building walls, is used to connect the personal computer
1 at someone's desk to the file server 2 in the telecommunications
room. The maximum length of cable 3 inside the wall cannot exceed
90 meters. Wall jack connectors 4 are used to connect the cords 5
from the computer and file server to the LAN cabling.
[0003] Cabling: Cabling is an important word in the term LAN
cabling because cabling includes the connectors 4 placed on the LAN
cable as well as the cable 3 itself. Thus, the performance of the
LAN cabling depends upon the connectors as well as the cable.
[0004] Installation: Technicians install the LAN cabling as a part
of new construction or as part of a LAN performance upgrade in
existing structures. In either case, the technicians pull the LAN
cable 3 through the walls and then place the connecting jacks 4 on
the ends of the cable. The jacks are then snapped into the wall
jack mounting plate and the installation is complete.
[0005] However, the technician is then required to test each LAN
cabling run or link with calibrated test equipment. This testing
certifies to the general contractor that the cabling run has been
correctly installed from the standpoint of signal integrity.
Hand-held LAN testers are used to perform these tests. The testers
drive the cabling with a series of different signal types and from
measurements of the received signals, determine if the cabling is
capable of supporting the telecommunication signals at the
prescribed data rate.
[0006] The LAN testers record the results of each test and, at a
later time, print out a test document indicating that the link
passed or failed. The technician gets paid for the links that pass.
If there are links that fail, the technician must re-test, and
often replace connectors that have been incorrectly or improperly
installed. The technicians keep testing and repairing the links
until they all pass.
[0007] LAN Testers: LAN testers are fairly sophisticated hand-held
test systems, which can test LAN links with a series of tests
covering a frequency range of 1 to 250 MHz, in the case of TIA
Category 6 cabling. FIG. 2 shows a typical LAN tester 6, with a
test adapter circuit board 7 connected to the LAN tester. The test
adapter circuit board includes a test jack connector 8. The purpose
of this test adapter is to provide a connection interface between
the LAN tester and the LAN link to be tested.
[0008] The test jack 8 allows the LAN tester 6 to connect to the
LAN link with a patch cord 9, as shown in FIGS. 3 and 4. Typical
lengths for patch cords are two meters, or approximately six feet.
This length allows the technician to conveniently connect the LAN
tester to the wall jacks 4 during test runs.
[0009] Standards: Technicians test their installed links with
reference to telecommunication industry standards. In the United
States the standard is specified by the TIA or Telecommunications
Industry Association. In Europe the standard comes from ISO, or
International Standards Organization. When testing a link, the
technician selects which type of link is being tested and the
corresponding sets of measurement limits, whether from TIA or
ISO.
[0010] The link is tested and the measured results are compared to
limits from the specified standard. If no limits are exceeded, the
link passes. If not, then the link fails and the technician must
work on the failed link, as required, until it passes. Often this
means reinstalling the connectors on the ends of the cable.
[0011] Standard Link Definitions: FIG. 5 shows the standard
permanent link, in simplified form, with 90 meters of LAN cable,
running within a structure's wall, or overhead in the ceiling. The
wall jacks, attached to the cabling ends, are used to connect the
link with equipment in the telecommunications room and to
individual items such as computers or printers within the office's
local area network. The TIA and ISO specify the length of 90 meters
as the maximum length for the permanent link.
[0012] Link Testing: FIG. 6 illustrates how the LAN testers check
the performance of a link. When testing a link (a procedure known
in the industry as "shooting" a link), two LAN testers are required
as shown. The technician connects a display end LAN tester 6A at
one end of the link, and the remote end LAN tester 6B at the other
end of the link. Since the display end LAN tester has a display
screen to show the measurement test results, the technician shoots
the link from the display end, controlling the test from there, and
viewing the test results.
[0013] During the test, first one unit applies test signals to one
end of the link while both units measure the results. Then the
roles are reversed with the signal application and signal
measurement taking place at the opposite ends of the link. When the
test is complete, the remote unit sends its data measurement files
to the display unit for final processing and storage within the
display unit. The limits for each test, specified by the selected
standard, are applied to the measurement data set to determine if
the link passed or failed the certification test.
[0014] Standard Links: Both the TIA and ISO have defined two types
of LAN links, the channel link and the permanent link. Each link is
shown and discussed below.
[0015] Channel Link: The channel link includes the LAN link and the
patch cords, as shown in FIG. 7, but does not include the
connection to the channel test adapter boards 7A. The channel link
measurement path includes the link 3 inside the walls, the mated
connector pairs at the walls and the patch cords and is supposed to
represent the performance of the final, complete telecommunications
link, which also uses patch cords to connect the personal computers
and file servers to each other. Since there is a longer length of
cabling in this path, the test limits for the channel link are not
as stringent as those for the permanent link.
[0016] Permanent Link: The permanent link includes the link 3, plus
the mated connector pairs at the wall jack, but it does not include
the patch cord, as shown in FIG. 8. Nor does it include the
connection to the permanent link test adapter board 7B. The
permanent link test evaluates only the cable within the walls, the
connector jacks at the wall, the plugs that are inserted into the
jacks, and two centimeters of cable that is attached to each of the
plugs. The permanent link test essentially represents the
performance of just the link cabling within the walls.
Consequently, the permanent link test limits are the tightest
measurement limits to pass.
[0017] As a result, technicians are often told that if their link
fails the permanent link test, to change over the LAN tester limits
to channel link limits and re-test. If the channel test passes, the
link may then be considered to pass under these conditions.
[0018] Consideration will now be given to the test issues faced by
the technicians as they test their installed Local Area Network
(LAN) cabling for compliance with the appropriate TIA or ISO
measurement test limits. The technician will certify the installed
link to either permanent or channel link measurement limits. It is
assumed that the technician has performed steps necessary to
calibrate the test equipment in the field before the LAN
certification test to assure maximum LAN tester measurement
accuracy.
[0019] Permanent Link Testing Issues
[0020] 1. Permanent Link Adapter Construction: Note the prior art
permanent link test adapters 7B shown in FIG. 8. Keep in mind the
permanent link comprises the cable in the wall plus the mated
connector pair at the wall jacks, but it does not include most of
the patch cord. The permanent link adapters (PLA's) are typically
fabricated by cutting a patch cord in half, and then soldering each
of the cut patch cord ends to a printed circuit board (PCB) within
the permanent link test adapter housing. These PCB's are designed
to cause very little signal integrity problems so that their
effects are ignored.
[0021] 2. Permanent Link Testing Lifetime: Permanent link adapters
have a limited test lifetime due to mechanical flexing of the patch
cord as it enters the PLA housing. When the patch cord has been
flexed beyond its maximum number of flexures, it will require
replacement. When this happens, the entire PLA has to be replaced.
In addition, for maximum test accuracy, both PLA's, the one at the
display end and the one at the remote end should be replaced.
[0022] 3. Dedicated PLA: The LAN testers often use a dedicated PLA
for each permanent link tested. This is because the circuit and
transmission line properties of the patch cord can be an important
part of the overall PLA measurement result. The installation
technician needs to be aware of what link he or she is testing, who
made the cabling, and what is the preferred type of PLA to use.
[0023] 4. Matched PLA Sets: Usually the technician will use a set
of PLA's matched to the cable type, by vendor, which is used in the
link. If the link is made with cabling, (that is, cable plus
connectors), from Vendor X, then a PLA made from Vendor X patch
cords will be used for the certification test.
[0024] 5. PLA Cost: The PLA's can be a costly item for the
installers, often $400 or more for a set of two. If the LAN cabling
installation testing company has several installers, each requiring
several different sets of vendor specific PLA's, this overhead item
can be rather costly. The cost comes from a dedicated printed
circuit board, within a plastic housing, to form the structure of
the PLA, which connects to the LAN tester.
[0025] 6. PLA Cross-talk: In addition, as LAN certification moves
to frequencies above 250 MHz, the performance of the PLA's as a
part of the measurement system becomes more critical. The measured
cross talk or lack of isolation between conductor pairs within the
PLA connection circuit board becomes a serious issue as frequencies
increase. When the isolation degrades beyond a certain level, the
LAN tester cannot measure the cabling pair-to-pair isolation
because it cannot "see" past its own PLA generated crosstalk.
[0026] The present invention provides the solution to this problem.
The solution is to use a connector with proven isolation properties
on the test adapter board, and then to connect to that test adapter
board with a patch cord having a connector which mates to the
connector on the adapter board.
[0027] 7. PLA Reference Plane Calibration: The last issue with
permanent link adapters is that of the measurement reference plane
location. The purpose of permanent link calibration is to refer all
permanent link measurements to a known point along the patch cord.
In particular, the permanent link measurement reference plane is
calculated to set this point at the end of the patch cord, 2
centimeters from the wall jack. From this calibration, all effects
from the patch cord are removed from the permanent link
measurement. The calibration procedure used to define and set this
reference plane at this point can involve taking an initial set of
permanent link calibration data and finally referring it to this
desired reference plane.
[0028] Channel Link Testing Issues
[0029] 1. Channel Link Adapters: Note the channel link test
adapters 7A shown in FIG. 7. Keep in mind the channel link includes
the link (i.e., the cable in the wall plus the mated connector
pairs at the wall jacks) and the patch cords but it does not
include either the plugs or the jacks at the channel test adapter
boards. The channel link adapters (CLA's) are fabricated by placing
a right-angle connector with appropriate isolation on the printed
circuit board mounted within the CLA housing. The right angle
connector is selected to provide significant pair-to-pair isolation
when mated with the patch cord used for the channel link
certification.
[0030] 2. CLA Testing Lifetime: Channel link adapters have a much
longer test lifetime when compared to permanent link test adapters
since the use of low cost replaceable patch cords solves the patch
cord mechanical flexure problem. The connector mounted on the
printed circuit board inside the CLA eventually wears out as the
cladding on the contacts wears off. Nevertheless, the testing
lifetime for the channel link adapter is considerably longer than
that for the permanent link adapter.
[0031] 3. Dedicated CLA: The LAN testers also use a dedicated CLA
when testing channel links since low cross talk, high isolation
connectors 8 are used on the channel link adapter printed circuit
board.
[0032] 4. Matched CLA Sets: Matched CLA sets are used by definition
by virtue of the high isolation right angle printed circuit board
connectors mounted on the PCB within the CLA housing. However, when
compared to the PLA, any type of patch cord can be used with the
CLA, so long as the patch cord is compliant with the cabling
category used for the link under test.
[0033] 5. CLA Cost: The CLA's are less costly than PLA's, since
they can use any compliant patch cord to connect to and test the
channel link.
[0034] 6. CLA Cross-talk: The channel link pair-to-pair isolation
is superior to that of the permanent link by virtue of the low
crosstalk connector used within the CLA module housing.
[0035] 7. CLA Reference Plane Calibration: The last issue with
channel link adapters is also that of the measurement reference
plane location. In particular, the channel link measurement
reference plane is set at the end of the patch cord connector right
at the input end of the patch cord, as shown in FIG. 7. With this
calibration, all effects from the patch cord input connector (i.e.,
the plug at the tester end) are removed from the channel link
measurement.
[0036] LAN Link Measurement Issue Summary
[0037] From the preceding discussion, when compared to channel link
adapters, permanent link measurements require the use of a separate
set of permanent link adapters, which add an undesirable set of
costs in terms of: 1) the permanent link adapters themselves; 2)
the number of dedicated PLA sets; and 3) limited PLA test lifetime
due to patch cord flexure failure. Permanent link adapters also
have more problems with minimizing pair-to-pair crosstalk when
compared to channel link adapters.
SUMMARY OF THE INVENTION
[0038] For these reasons, in the present invention a
calibration/measurement method is proposed the objectives of which
are to:
[0039] 1. eliminate completely the permanent link test adapter;
[0040] 2. reduce LAN measurement overhead support costs;
[0041] 3. improve signal integrity;
[0042] 4. increase LAN link measurement accuracy at frequencies
above 300 MHz; and
[0043] 5. provide a means to measure permanent links using channel
adapters and low cost patch cords.
[0044] Phase
[0045] Preparatory to a description of the method of the present
invention, a discussion of phase needs to be presented. The
capability of phase measurement is a key attribute of the LAN
tester of the present invention. That is, in addition to magnitude,
the hand-held LAN tester of this invention can measure phase. This
capability permits the tester to set a measurement reference plane
at one specified point along the LAN link to be measured. The
original calibration reference plane may be set at a point along
the link at a point, which is easy to set, measure, define and
implement.
[0046] Phase also allows the tester to easily move this original
calibration reference plane and all of its associated LAN link
measurements to another, new, reference plane location at any time
during the LAN link testing. Specifically, with phase information,
a display end and remote end can each move the phase reference
plane from within the channel link adapter printed circuit board,
through the mated pair of connectors at the CLA output and anywhere
down the length of the patch cord, and up to the mated pair of
connectors at the wall jack, in any of the four possible locations
as shown in FIG. 9. Movement of the phase reference plane enables
the tester of this invention to use a channel link adapter and low
cost patch cord to perform permanent link measurements.
[0047] In brief, the method involves the calibration step of
measuring the overall scattering parameters S.sub.T for each of the
patch cords plus mated connectors pairs at each end of the patch
cords, as indicated in FIG. 10. The scattering parameters S.sub.B
of each patch cord can be obtained from known characteristics of
the cord. This, together with the total scattering parameter matrix
S.sub.T allows calculation of the scattering parameters S.sub.A and
S.sub.C of the mated connector pairs at the ends of the cords. With
the scattering matrices of the mated connector pairs S.sub.A and
S.sub.C and the patch cord S.sub.B known, the reference plane may
be moved anywhere along the cord from within the LAN tester to
perform either permanent link or channel link tests.
BRIEF DESCRIPTION OF THE DRAWINGS
[0048] FIG. 1 is a diagrammatic sketch of a LAN cabling connection
from a work area to a telecommunications room.
[0049] FIG. 2 is a diagram of a prior art LAN tester with a test
adapter and test jack.
[0050] FIGS. 3 and 4 illustrate a prior art LAN tester connection
with a patch cord.
[0051] FIG. 5 illustrates a standard 90 meter link.
[0052] FIG. 6 illustrates the process for testing or "shooting" a
link with LAN testers.
[0053] FIG. 7 illustrates a channel link configuration.
[0054] FIG. 8 illustrates a permanent link configuration.
[0055] FIG. 9 illustrates movement of the measurement reference
plane with phase, as taught by the present invention.
[0056] FIG. 10 illustrates the LAN testers of the present
invention.
[0057] FIG. 11 is a plot of drive signal and a resulting signal
measured by the LAN tester of the present invention.
[0058] FIG. 12 is an schematic diagram of the phase measurement
circuit in a display unit of the present invention.
[0059] FIG. 13 is an illustration of setting the measurement
reference plane during factory calibration, according to the
present invention.
[0060] FIG. 14 is an illustration of movement of the reference
plane through a mated connector pair.
[0061] FIG. 15 is an illustration of movement of the reference
plane down the patch cord, according to the present invention.
[0062] FIG. 16 illustrates how the reference plane at point 2 of
FIG. 9 relates to the reference plane at point 3 of FIG. 9.
[0063] FIG. 17 is an exploded perspective view of the LAN tester
display unit of the present invention.
[0064] FIG. 18 is an exploded perspective view of the underside of
the tester unit.
[0065] FIG. 19 is a block diagram of the digital control circuit
board of a LAN tester unit of the present invention.
[0066] FIG. 20 is a block diagram of the analog circuit board of
the present invention.
[0067] FIG. 21 is a detailed phase measurement block diagram of the
present invention.
DETAILED DESCRIPTION OF THE INVENTION
[0068] A schematic representation of the LAN testing system of the
present invention is shown in FIG. 9. The testing systems includes
a hand-held display unit 10, a hand-held remote unit 12 and first
and second patch cords 14 and 16. Each patch cord comprises a first
plug 14A, 16A at one end, the actual cable 14B, 16B and a second
plug 14C, 16C at the other end. The display unit 10 has a channel
link adapter board 18 on which is mounted a first connector jack
20. The jack is exposed to the exterior of the display unit. Jack
20 can receive the plug 14A or 16A of a patch cord to form a first
mated connector pair. When shooting a link, the other plug 14C, 16C
of the patch cord mates with a wall jack 22 attached to the link 24
running inside the walls. The remote unit 14 similarly has a
channel link adapter board 26 on which is mounted a second
connector jack 28. Both of the connectors 20 and 28 are preferably
right-angle connectors with appropriate pair-to-pair isolation. An
RJ-45 jack or a Siemon terra jack for higher frequencies are
suitable. Jack 28 receives the plug 16A of the second patch cord to
form a second mated connector pair. When shooting a link, plug 16C
of the second patch cord 16 connects to a wall jack 30 on the end
of the link 24. The display and remote units contain appropriate
radio frequency and electronic circuitry for testing the link. The
display unit also has user-actuated switches for starting and
controlling the testing functions, as well as a display that
communicates to the user whatever data is appropriate. The display
unit also has a computer processor for performing the calculations
described below, and memory to store measured scattering parameters
and other data.
[0069] Operation of the LAN testing system is as follows. First a
field calibration with the display and remote units and both patch
cords must be performed. The object of this calibration is to set a
measurement reference plane for the display unit and the remote
unit by using any two patch cords with a set of channel link
adapters connected to the display and remote units as shown in FIG.
10. The two patch cords should be made by the same vendor with
identical plugs on each end, but they do not have to be the same
length.
[0070] Scattering Parameters
[0071] Since the display and remote units can measure phase, the
complete patch cord consisting of the patch cord plugs and the
patch cord itself can be measured or characterized by measuring
their frequency response with scattering, or [S] parameters. From
factory calibration, the measurement reference plane on the channel
adapter printed circuit board will be at the input to the
right-angle connector jacks 20, 28 on the channel link adapter
boards 18, 26.
[0072] Measurement Steps:
[0073] 1. Connect patch cord 14 between the two units.
[0074] 2. Measure all four scattering parameters of the first patch
cord 14, so connected, including the mated connector pairs 20,14A
and 28,14C at each channel link adapter board 18,26.
[0075] 3. Save the total, measured scattering data [S.sub.T].sub.1
for the first patch cord 14
[0076] 4. Connect the second patch cord 16 between the two
units.
[0077] 5. Measure all four scattering parameters of the second
patch cord 16, so connected, including the mated connector pairs
20,16C and 28,16A at each channel link printed circuit board 18,
26.
[0078] 6. Save the total, measured scattering data [S.sub.T].sub.2
for the second patch cord 16.
[0079] Calculation Steps
[0080] 1. The elements for the scattering matrix, for each of the
patch cords, are a set of simple equations or terms, with known
formulation as follows:
[0081] As a 2-port example, consider: j:={square root}{square root
over (-1)} c:=3.multidot.10.sup.8 M/second
[0082] Assume Values for Input Mated LAN Connector Pair A
Matrix--[S.sub.A] 1 SA 1 , 1 := 0.0400 + 0.01 j SA 1 , 2 := 0.3 -
0.1 j SA 2 , 1 := SA 1 , 2 SA 2 , 2 := SA 1 , 1 _ SA 2 , 2 = 0.04 -
0.01 i SA := ( SA 1 , 1 SA 1 , 2 SA 2 , 1 SA 2 , 2 ) SA = ( 0.04 +
0.01 i 0.3 - 0.1 i 0.3 - 0.1 i 0.04 - 0.01 i ) det_A := SA 1 , 1 SA
2 , 2 - SA 1 , 2 SA 2 , 1
[0083] Patch Cord Matrix--[S.sub.B] (Assume that the Line is a
Perfect Match) 2 L := 2 Meters Assume F = 600 MHz , NVP = 0.75 :=
0.002 F := 600 10 6 NVP := 0.75 := 2 F c NVP := + j := L = 4
.times. 10 - 3 + 33.51 i SB 11 := 0 SB 12 := - SB 21 := - SB 22 :=
0 SB := ( SB 11 SB 12 SB 21 SB 22 ) SB = ( 0 - 0.498 - 0.863 i -
0.498 - 0.863 i 0 ) det_B := SB 1 , 1 SB 2 , 2 - SB 1 , 2 SB 2 ,
1
[0084] Output Mated LAN Connector Pair A Matrix--[S.sub.C], (Note
Relationships to [S.sub.A]) 3 SC 1 , 1 := SA 2 , 2 SC 1 , 2 := SA 2
, 1 SC 2 , 1 := SA 1 , 2 SC 2 , 2 := SA 1 , 1 SC := ( SC 1 , 1 SC 1
, 2 SC 2 , 1 SC 2 , 2 ) SA = ( 0.04 + 0.01 i 0.3 - 0.1 i 0.3 - 0.1
i 0.04 - 0.01 i ) for Reference SC = ( 0.04 - 0.01 i 0.3 - 0.1 i
0.3 - 0.1 i 0.04 + 0.01 i ) det_C := SC 1 , 1 SC 2 , 2 - SC 1 , 2
SC 2 , 1
[0085] 2. To an acceptable degree of accuracy, the patch cord
characteristic impedance Zo is known, and to a very good, first
order approximation, may be considered to be Zo=100 Ohms.
[0086] 3. The electrical length of the patch cord will be known.
The length may be specified by the manufacturer of the tester
units, or it can be measured by the LAN tester.
[0087] 4. To an acceptable degree of accuracy, the scattering
matrix for the mated jack and plug at each end of the patch cord
can be assumed to be identical.
[0088] 5. Then, using the justifiable assumptions, 1-4 above, and
[S.sub.T].sub.1, the measured total scattering matrix for the first
patch cord 14, the scattering matrix for the mated jack and plug
pair at each end of the patch cord can solved for.
[0089] 6. With the mated connector pair scattering matrix and the
scattering matrix for patch cord 14, the measurement reference
plane may be moved through the mated connector pair on the printed
circuit board. This reference plane location is necessary to
perform a channel link test; or the reference plane may be moved
further down the patch cord to within 1 or 2 centimeters of the
wall jack, in order to perform a permanent link measurement.
[0090] 7. The same set of measurements and calculations are then
made using the second patch cord 16.
[0091] 8. The scattering parameters for the mated connector pairs
are saved for testing throughout the day or until another patch
cord set is selected, at which time the field calibration procedure
is repeated.
[0092] The scattering parameter matrices can be manipulated using
linear algebraic calculations to solve for the elements of the
mated LAN channel connector scattering matrix. In this set of
calculations, a set of scattering parameters for the mated
connector pair is assumed, and following established formulation,
the complete, total scattering matrix [S.sub.T] is calculated by
combining the scattering matrices of the mated connector pair with
that of the patch cord transmission line.
[0093] Then, with [S.sub.T] as the "given" final, measured result,
and with the assumptions for the patch cord transmission line,
assumptions 2 and 3 above, plus assumption 4 which assumes
identical scattering matrices for the two mated connector pairs,
the program solves for the elements of the mated connector pair
scattering matrix [S.sub.A].
[0094] The program solves for and calculates the same values for
[S.sub.A] as were assumed in the original calculation for the total
[S.sub.T] matrix. This calculation confirms the mathematical model
as correct.
[0095] Turning now to a consideration of the phase measurement
aspect of the invention, the LAN tester of the present invention
measures the relationship between two signals as it tests LAN
cabling for compliance with published LAN cabling performance
standards. The signal relationships measured by the tester are
ratios in magnitude, and include the phase relationship between the
two signals. Note that this phase measurement under discussion is
the phase between a drive signal voltage and the corresponding
coupled or reflected voltage due to that same drive signal. These
two signals are measured at a specified reference plane determined
by factory or field calibration procedures.
[0096] Phase difference can be shown between two sinusoidal signals
at the same frequency. In the plot shown in FIG. 11, the V_Drive
trace (the solid line) corresponds to the drive signal into the LAN
cabling. The V_Meas trace (the dotted line) is the resulting signal
to be measured by the LAN tester. Note that the amplitude of V_Meas
is 40% of the amplitude of V_Drive. V_Meas also lags V_Drive by 30
degrees of phase. This lagging phase relationship between V_Drive
and V_Meas can also be seen in the plot.
[0097] If the ratio of V_Meas to V_Drive is calculated, one then
calculates, for example, the crosstalk term relating the drive
signal on one LAN cable pair and the coupled, crosstalk signal
which appears on another LAN conductor pair. When the ratio,
V_R=V_Meas/V_Drive is calculated, the .vertline.V_R.vertline., the
magnitude of
V_R=.vertline.V_R.vertline.=.vertline.V_Meas.vertline./.vertline.V_Drive.-
vertline.=0.4/1.0=0.4. Thus .vertline.V_R.vertline.=0.4.
[0098] The phase between the two signals must be calculated using
one of the signals for the phase reference. In this case, the
V_Drive signal is defined to be the reference signal. The phase
relationship of V_Meas is then said to lag the reference signal,
V_Drive, by 30 degrees of phase. Since a phase angle is involved,
the ratio, V_R=V_Meas/V_Drive is a complex number, with a
corresponding magnitude, .vertline.V_R.vertline. and phase angle,
.phi..sub.--=-30 degrees. The negative sign on .phi._R indicates
that V_Meas lags V_Drive by 30 degrees in phase. Thus V_R=0.4/-30
degrees.
[0099] Phase may also be calculated from the time relationship of
two square wave signals, by computing the time difference between
two corresponding edges of the square wave signals. This is
illustrated in FIG. 12 where the signals travel from left to right.
Note in FIG. 12 the two square waves, V_Meas and V_Drive, where the
leading edge of V_Meas lags the leading edge of the reference
V_Drive square wave by the time difference, .DELTA.t. This time
difference may be used to calculate the phase between the two
signals, by relating .DELTA.t to the period, T.sub.Clock, of a
precise reference clock running at frequency, F.sub.Clock.
T.sub.Clock=1/F.sub.Clock
[0100] The phase in degrees, .phi._R, between these two square
waves, is then:
.phi..sub.--R=360.times.(.DELTA.t/T.sub.Clock) degrees
[0101] The phase measurement circuit determines the value for
.DELTA.t, and outputs a signal related to the phase between the two
square waves, V_Meas and V_Drive. The LAN tester of the present
invention uses a programmable gate array to measure .DELTA.t.
[0102] The LAN tester can measure phase, which needs to be
referenced to a measurement reference plane, as discussed
below.
[0103] 1. Set the Measurement Reference Plane Initially--During
calibration, the phase measurement capability permits the LAN
tester to set, or define a measurement reference plane at one
specified point along the LAN link to be measured. This reference
plane, defined during the factory calibration procedure, may be set
anywhere along the link at any point, to permit measurements which
are simple and convenient to make. The calibration procedure is
shown in FIG. 13.
[0104] The reference plane location is defined or set during
initial factory calibration at the display and remote ends with the
procedure shown in FIG. 13. Plugs containing short-circuit,
open-circuit and terminations are applied in sequence to the jack
on the channel link adapter. Swept frequency measurements are taken
with each plug connected to the jack. From the measured data, which
includes phase information, the display or remote end sets its
reference plane at the point looking into the jack on the CLA
printed circuit board shown with the dotted line. With this
reference plane set at this point, phase information also allows it
to be moved from this point, up and down the patch cord.
[0105] 2. Move the Measurement Reference Plane--Following patch
cord field calibration, phase also allows the LAN tester to easily
move this original calibration reference plane, during link
testing. Phase allows the original reference plane to be moved to a
new reference plane location at any time during the LAN link
testing. Specifically, with phase information, a display end,
and/or remote end can each move their phase reference plane from
within the channel link adapter PCB, through the mated pair of
connectors at the CLA output, anywhere down the length of the patch
cord and up to the mated pair of connectors at the wall jack, in
any of the four possible locations as shown in FIG. 9.
[0106] 2a. Reference plane movement through the mated connector
pair from 1 to 2, shown in FIG. 14, on the channel link adapter
module is performed using the [S.sub.21].sub.Connector Pair data
measured and calculated for the mated connector pair during patch
cord field calibration. This step sets the reference plane for
channel link testing.
[0107] 2b. Reference plane movement down the patch cord from 2 to
3, shown in FIG. 15, moving down the patch cord is performed using
the [S]--parameter data measured and calculated for the patch cord
during field calibration.
[0108] Note that the desired length down the patch cord,
L.sub.Line, from 2 to 3 is expressed in the physical length units
of inches. L.sub.Line, in inches, needs to be converted into
equivalent electrical phase length, .phi..sub.Line in degrees.
[0109] During the patch cord field calibration, the NVP (nominal
velocity of propagation) for the patch cord is determined through
measurement. From this value, the corresponding electrical phase
length, .phi..sub.Line, moving from 2 to 3 is calculated using:
.beta.=(360.times.f)/(NVP.times.c) degrees/inch
[0110] where:
[0111] c=velocity of light in freespace=1.1811.times.10.sup.10
inches/second
[0112] f=signal frequency in Hertz
[0113] Then .phi..sub.Line in degrees is calculated using:
.phi..sub.Line=L.sub.Line.times.(360.times.f)/(NVP.times.c)
degrees
[0114] Relating this formulation to just the patch cord reference
planes, moving from 2 to 3 can be seen in FIG. 16.
[0115] The measured LAN cable data is moved thru the mated
connector pair from 1 to 2 using the [S.sub.21].sub.Connector Pair
data as shown in FIG. 14. With reference to FIG. 15, which shows
how a patch cord length in inches is related to the equivalent
patch cord electrical length in degrees, FIG. 16 relates these
terms per the formulation below: 4 [ S ] Patchcord = [ S 11 P S 12
P S 21 P S 22 P ]
[0116] For a reasonably well-matched patchcord, this expression
becomes: 5 [ S ] Patchcord = [ 0 - jLine - jLine 0 ]
[0117] If the patch cord has characteristic impedance, Z Op, not
equal to Z.sub.0=100 Ohms, then the patch cord S.sub.11p and
S.sub.22p are non-zero and are then replaced by non-zero values
calculated by using standard transmission line theory.
[0118] Finally, the measured LAN cable data with reference to plane
2, is related to plane 3, through the use of, [S].sub.Patchcord,
the patch cord scattering matrix. 6 [ S ] Patchcord = [ 0 - jLine -
jLine 0 ]
[0119] From this matrix it can be seen that for a well-matched
patch cord, the usual case, there is no effect upon the S.sub.11
and S.sub.22 mated connector pair terms. The only effect to the
mated connector pair scattering matrix is the added phase term,
e.sup.-j.phi.Line.
[0120] Thus, with the patch cord NVP known by measurement or
specification, the measurement reference plane may be moved through
the mated connector pair on the channel link adapter board and down
the patch cord a specified number of inches from plane 2, at the
output of the mated connector pair.
[0121] Turning now to a more detailed description of the tester
units, the exploded views of FIGS. 17 and 18 show the overall
physical configuration of the tester and show how its printed
circuit boards are housed. The tester shown is a display unit 10.
It will be understood that the remote unit is similar. The tester
has a housing including a front enclosure 32 and a rear enclosure
34. The rear enclosure defines a receptacle or well 36 for
receiving and mounting a channel link adapter printed circuit board
37. Inside the housing there is a digital control module 38 that
drives and controls an analog stimulus/measurement module 40. Both
modules are built into printed circuit boards and will be referred
to herein as the digital board and the analog board. The analog
board 40 includes a connector 42 on the underside thereof. This
connector is releasably engageable with a mating connector on the
channel link adapter through an opening 43 in the rear enclosure at
the bottom of the well 36. A time domain reflectometer (TDR) 44
measurement capability is provided by a third separate module.
Other components shown in FIG. 17 include a PCMCIA card holder 46,
a universal serial bus (USB) port 48 and a serial port 50. These
are mounted on the digital board 38. A color display unit 52 and a
keyboard 54 are mounted on or in the front enclosure 32. Further
details of the physical arrangement of the housing may be as shown
and described in U.S. patent application Ser. No. 09/863,810, filed
May 22, 2001 entitled "Apparatus with Interchangeable Modules for
Measuring Characteristics of Cables and Networks", the disclosure
of which is incorporated herein by reference.
[0122] The overall function of the digital control module 38 is
shown in the digital control circuit block diagram of FIG. 19. The
digital board is controlled by a high-speed central processing unit
(CPU) 56 driven by the firmware installed within the tester.
Several memory blocks (not shown) may be provided, as well as a RAM
memory 58, a small boot flash memory 60, and a larger boot flash
memory 62. The tester communicates with an external personal
computer (PC) 1 either by using the USB 48, or with a serial
interface connection 50 to the CPU 56. Flash memory or networking
cards 64 can be installed in the tester, which connect to the CPU
through the PCMCIA block 46. These cards can be used to store
additional test results, or to upload new firmware to the CPU.
Other connections to or from the CPU include the keyboard 54, the
color display 52, a speaker phone 66, a real time clock 68 and
temperature sensors 70 to compensate the analog board performance
as temperature rises.
[0123] Of utmost importance is communication with the analog board
40 through the I/O bus 72. This bus is shown as a separate block
because it interfaces control commands to the analog board 40 from
the digital board 38, and it returns measured data from the analog
board for storage in the display unit memory and for display on the
color display 52.
[0124] The LAN tester analog circuit block diagram of FIG. 20 shows
the major functional blocks on the analog board 40. Other blocks
have been omitted for clarity. The analog board generates a set of
continuously varying low frequency (LF) and radio frequency (RF)
signals which are applied to one selected conductor pair of the LAN
cabling through the use of signal switching relay banks 74 on the
analog board. The same relay banks carry the return signal to be
measured from another selected LAN cable conductor pair back into
the analog board. Circuit blocks on the analog board then condition
the return test signal and measure its characteristics relative to
the applied drive signal. The low frequency LF measurements include
cable capacitance, length, conductor wire DC resistance, wire
mapping and delay. The circuit blocks associated with these lower
frequency signals are identified by the associated notation.
[0125] Note the notation of "MUX` several places in the block
diagram. A MUX is shorthand notation for a multiplexer, which is a
switching device that routes an input of several different signals
to a selected signal path. The LAN tester analog board uses several
MUX's since it is a four-channel test instrument, capable of
testing two of the possible four conductor pairs of the LAN cable
under test. The MUX's are required for signal routing and channel
to channel signal isolation.
[0126] Circuit blocks 76, 78 are shown for RS-485 blocks, one for
communication, and another for LAN tester interface with the gate
array and measurement IC 80, and for DC power control and power
management on the analog board.
[0127] The analog board 40 also measures the RF parameters of cable
crosstalk, return loss and attenuation. Specifically the analog
board measures the ratio of the amplitude of the returned signal
divided by the amplitude of the RF drive signal. Circuitry has been
added to the analog board in the unit to measure the phase of the
returned test signal relative to the RF drive signal sent out on
the selected conductor pair.
[0128] The RF drivers 82 send a signal from the RF synthesizer 84
out on one pair of the LAN cable conductors via the RF signal
switching relays 74. The drive signal is also sent to the
return-loss bridges 86.
[0129] The resulting test signal comes into the tester via the same
set of RF relays 74 and is routed through the return-loss bridges
86 to the RF mixer block 88. There it is mixed with the local
oscillator (LO) signal and converted into the test IF (intermediate
frequency) signal.
[0130] As mentioned above, the RF drive signal is also sent into
the return-loss bridges 86. As shown in FIG. 20, the drive signal
sent to the return-loss bridges enters the mixer 88 and is
converted to a phase reference IF signal. All IF signals associated
with the LAN measurement are compared with this phase reference IF
signal to determine the phase of that measurement signal.
[0131] Once both the Ref IF and Test IF signals have been created
they are delivered to the phase detector 90 and the reference and
test magnitude detector blocks 92 and 94. The phase detector block
90 sends the phase information into the gate array and measurement
IC 80. The outputs from the reference and test magnitude detectors
92, 94 are sent to the analog-to-digital (A to D) Mux 96 and then
to the A to D converter 98. From there the magnitude ratio signal
is sent to the gate array and measurement IC (integrated circuit)
80.
[0132] The gate array and measurement IC 80 finishes the
computation of the phase between the test and reference IF signals,
and the ratio of their amplitudes, to formulate a complex number
representation of the measurement. Output from the IC 80 is placed
on the analog I/O bus 72, which communicates with the digital board
38. Thus, the phase measurement function for the testing unit is
controlled from the digital board, but is measured and computed on
the analog board. The measurement results are then carried to the
digital board from the analog board.
[0133] The LAN tester phase measurement block diagram of FIG. 21
shows this function on a phase measurement system level. This block
diagram also shows computation of the magnitude ratio on the analog
board. The I/O bus 72 carries the control signals from the digital
board to the analog board, and it also carries both the phase and
magnitude of the test signals to the digital board. Once the test
signal has been delivered to the display board it can be stored in
the memory or shown in color graphic form, plotted on the display
screen 52.
[0134] Regarding measurement speed, the tester architecture has
been designed in such a manner that a LAN cable conductor pair may
be driven with RF signals from either the display or remote test
units 10 or 12. All other non-driven lines may then be
simultaneously connected to the measurement circuitry via the MUX
circuitry on the analog boards within each unit. This design
feature provides for significantly reduced test times, while still
providing measurement of test signal magnitude and phase.
[0135] While a preferred form of the invention has been shown and
described, it will be realized that alterations and modifications
may be made thereto without departing from the scope of the
following claims.
* * * * *